Carbon Dioxide as a Carbon Resource – Recent Trends and Perspectives

Carbon Dioxide as a Carbon Resource – Recent Trends and Perspectives

Markus H¨olscher^a, Christoph G¨urtler^b, Wilhelm Keim^a, Thomas E. M¨uller^d, Martina Peters^c, and Walter Leitner^a

aInstitut f¨ur Technische und Makromolekulare Chemie, RWTH Aachen University, Worringerweg 1, 52074 Aachen, Germany

bBayerMaterialScience, 51368 Leverkusen, Germany

cBayerTechnologyServices, 51368 Leverkusen, Germany

dCAT Catalytic Center, Institut f¨ur Technische und Makromolekulare Chemie, RWTH Aachen University, Worringerweg 1, 52074 Aachen, Germany

Reprint requests to Prof. Dr. Walter Leitner. Fax: Int. + 241 8022177.

E-mail:leitner@itmc.rwth-aachen.de

Z. Naturforsch.2012,67b, 961 – 975 / DOI: 10.5560/ZNB.2012-0219 Received August 15, 2012

Dedicated to Professor Heribert Offermanns on the occasion of his 75^thbirthday

With the growing perception of industrialized societies that fossil raw materials are limited re- sources, academic chemical research and chemical industry have started to introduce novel catalytic technologies which aim at the development of economically competitive processes relying much more strongly on the use of alternative carbon feedstocks. Great interest is given world-wide to car- bon dioxide (CO2) as it is part of the global carbon cycle, nontoxic, easily available in sufficient quantities anywhere in the industrialized world, and can be managed technically with ease, and at low cost. In principle carbon dioxide can be used to generate a large variety of synthetic products ranging from bulk chemicals like methanol and formic acid, through polymeric materials, to fine chemicals like aromatic acids useful in the pharmaceutical industry. Owing to the high thermody- namic stability of CO2, the energy constraints of chemical reactions have to be carefully analyzed to select promising processes. Furthermore, the high kinetic barriers for incorporation of CO₂into C–H or C–C bond forming reactions require that any novel transformation of CO2must inevitably be as- sociated with a novel catalytic technology. This short review comprises a selection of the most recent academic and industrial research developments mainly with regard to innovations in CO₂chemistry in the field of homogeneous catalysis and processes.

Key words:CO2, Homogeneous Catalysis, Renewable Energy, Methanol, Formic Acid

Table of Contents 1. Introduction

2. Synthetic Transformations of CO₂ Using Metal Complexes as Catalysts

2.1 Reactivity of CO₂at metal centers

2.2 Oxidative coupling of CO₂ and olefins – metal- lalactones as intermediates for the catalytic synthesis of acrylic acid

2.3 Insertion of CO₂into M–O bonds – synthesis of carbonates, polycarbonates and polyurethanes 2.4 Insertion of CO₂into M–H bonds – synthesis of methanol and formic acid

2.4.1 Hydrogenation of CO2to methanol 2.4.2. Hydrogenation of CO₂to formic acid

2.5 Insertion of CO₂into the M–C bond – direct syn- thesis of aromatic carboxylic acids

3. Conclusion and Outlook 1. Introduction

For more than a century value creation in the or- ganic chemical industry has been based mostly on the steady and reliable supply of fossile carbon sources with petrochemistry dominating the last six decades.

A notable exception is the generation of certain basic

c

2012 Verlag der Zeitschrift f¨ur Naturforschung, T¨ubingen·http://znaturforsch.com

chemicals via biomass pyrolysis in parts of the Ger- man chemical industry in the early 20th century. In fact, the implementation of chemical technologies in the 20th century strongly reflected the general percep- tion that fossil resources will remain available in suffi- cient quantities as the most economically feasible op- tion. The same observation can be made for the pro- duction and usage of energy. A tremendous amount of fossil resources is currently used for the genera- tion of electricity and heat for private and industrial applications and to generate liquid fuels for transporta- tion systems. In 2011, ca.33 % of the global energy consumption was satisfied by oil, while the amount of non-energetic use of oil in North America and Eu- rope amounts toca.20 % only [1]. The total amount of carbon-based raw materials used in 2006 in Germany was equivalent to ca.498 million tons stone coal en- ergy equivalents from which onlyca.35 Mio t or 7 % were used non-energetically [2].

Inevitably, the use of fossil raw materials for en- ergy production leads to the formation of carbon diox- ide (CO₂). Even under the most optimistic scenarios for the replacement of fossil-based technologies, the above numbers clearly indicate that CO₂will be gen- erated as waste material from the energy sector in more than sufficient quantities to constitute an attractive car- bon source for the chemical supply chain in the fore- seeable future. The technologies to isolate CO₂ from such sources have reached a very mature state in the development of carbon capture and storage (CCS) [3].

It should be noted that also other highly concentrated streams of CO₂are available from industrial as well as even natural sources [4].

With regard to the correlation between a rising CO2concentration in the atmosphere and currently ob- served changes in the global climate it should be noted at this point that the increased usage of CO₂as a build- ing block for carbon-based compounds and materials cannot provide a ,,silver bullet“ to solve the problem of the world-wide CO₂emissions. The anthropogenic annual CO₂ emissions in 2010 were estimated to be 29 billion tons [5]. The current utilization of CO₂as an industry gas amounts toca.20 million tons while CO₂usage as a raw material for chemical production reachesca.110 million tons, largely dominated by the synthesis of urea as fertilizer [6–9]. The annual pro- duction of urea amounted to 153 million metric tons in 2011 with a usage of ca. 112 million tonnes of CO₂ as a feedstock [6]. The largest fraction of urea

is used as agricultural fertilizer, and smaller amounts are consumed for the production of fine chemicals, in polymer synthesis (e. g.as a starting material for urea- formaldehyde or urea-melamin resins) or for emis- sion control in automotives. Given the typical scale of chemical products, urea is clearly an exception, and the range of 10²−10³Mio tons defines more or less the up- per limit of the amount of CO₂that could possibly be exploited for chemical products.

Nevertheless, the increased valorization of CO₂can contribute significantly to the development of sustain- able production processes: The compound is the most abundant C1 building block, available in sufficient quantities practically anywhere in the industrialized world. It is non-toxic, and can be handled readily and safely on a technical scale. With the increasing price of other carbon sources and the recent improvement of CO2capture technologies, it is also becoming more and more economically attractive. This opens new op- portunities for the chemical synthesis of high-quality goods and can furthermore support the introduction of even safer production technologies in usually densely populated areas.

In this context it may be justified to explicitly name the at the time visionary publication by Friedrich Asinger (Chair of Technical Chemistry at RWTH Aachen University from 1959 to 1972), who pointed out already in 1986 that the use of CO₂as feedstock for the production of methanol could contribute sig- nificantly to an alternative way of storing energy and producing chemical goods [10]. These arguments were reiterated and strengthened by Keim and Offermans in 2010 [11]. Whereas the ecologic, economic and soci- etal drivers for the interest in CO₂utilization have seen various changes in the more than 25 years since then, the scientific challenges for chemistry and chemical engineering remain the same: To address the thermo- dynamic constraints and the kinetic limitations of the transformations of the carbon dioxide molecule!

Various research activities for CO₂utilization using catalytic technologies have been undertaken, and in the past years the field has seen re-newed dynamics result- ing in impressive progress based on the ever increasing insight into the molecular basis of catalysis and novel reaction engineering concepts [12–15]. In particular, homogeneous catalysts have played an important role ever since the early years of CO2chemistry.

The group of Wilhelm Keim at the Institut f¨ur Tech- nische Chemie und Petrolchemie at RWTH Aachen

University studied the telomerization of dienes with CO₂already in the mid 1970’s [16,17]. Stoichiomet- ric [18,19] as well as catalytic [20] reactions lead- ing to C–C bonds under CO₂ incorporation evolved in the following years. As a result of his early stud- ies in Aachen Arno Behr published a monograph that formed the leading reference in the field for many years to come [12]. Also the work of Dieter Vogt on the in- sertion of CO₂originated from this school [21].

Much research has been devoted to the synthesis of a variety of intermediate chemicals like acids, lactones, carbonates, as well as polymeric materials. Very re- cently a number of promising contributions were made in these areas. On the side of fundamental research, they address for example the direct synthesis of car- boxylic acids from CO₂, based on different mechanis- tic principles such as oxidative coupling or CO₂inser- tion to form new C–C bonds. Much closer to indus- trial application is the synthesis of polycarbonates and polyurethanes which incorporate substantial amounts of CO₂viathe formation of C–O bonds. Somewhere in between is the formation of C–H bonds by hydro- genation of CO₂, leading for example to methanol or formic acid as possible large-volume products. It is the aim of this short review to summarize some of the lat- est developments in homogeneous catalysis for the uti- lization of CO₂in the chemical value chain on the basis of representative examples from these areas.

2. Synthetic Transformations of CO₂Using Metal Complexes as Catalysts

2.1 Reactivity of CO₂at metal centers

Much pioneering work on catalytic CO2chemistry was conducted around the world including the groups of Behr [22–27], Hoberg [28–35], Walther [36,37], and Dinjus [38–42], as well as Inoue [43–47], Musco [48], Yamamoto [49,50], Darensbourg [51, 52], and others [53–57]. These efforts have devel- oped a fundamental understanding of the chemistry of metal complexes with CO₂ providing a sound basis for the rational development of catalytic transforma- tions of this raw material. A most recent overview by Rieger, Herrmann, and K¨uhn and coworkers [58], who focus on important developments of transition metal- based CO2chemistry, provides an excellent entry into the field. A review involving some of the authors of this manuscript [59] discusses the current technologi-

cal status and future trends for catalytic CO₂utilization in industry, including some basic criteria for the evalu- ation of the implementation of such strategies.

Considering solely the interaction of carbon dioxide with a metal center, CO₂can bind by three general co- ordination modes as depicted in Fig.1. In addition to the mononuclear binding modes shown here, multinu- clear complexes containing bridging CO₂ ligands are also known [14]. The ligand CO₂comprises a weakly Lewis acidic C atom and two weakly Lewis basic O atoms, and the C=O double bonds as possible coordi- native sites. Consequently, the electronic nature of the complex fragment will strongly influence the preferred binding mode. Electron-rich metal centers favor1and 2, which can be isolated as stable complexes in many cases. More electron-deficient centers will lead to 3, observed only as transient species.

Although coordinated CO2molecules can undergo a variety of transformations, their involvement in cat- alytic processes is not a necessary prerequisite. The most obvious possibility for C–E (E=C,H,O,N) bond formation in catalytic cycles is the formal insertion of CO₂into a M–E bond of a metal complex (Scheme1).

Alternatively, CO₂can be incorporated into an organic frameworkviaoxidative coupling with another unsat- urated ligand as exemplified for an olefin in Scheme1.

Whereas the coordination of both coupling partners is typically assumed in oxidative coupling processes, the insertion reactions show often lower barriers for path- ways without pre-coordination or only very weak in- teractions of CO₂at the metal center.

These principal reactions comprise the most abun- dant types of currently known transformations of CO₂ at metal complexes. In the following sections some reactions which would be very interesting to be put into industrial practice or are close to being realized

Fig. 1. Properties of CO₂as a ligand (top) and coordination types of CO2in metal complexes (bottom) [14].

Scheme 1. Reactions of CO₂with organometallic complexes:

a) Insertion of CO2into a metal-element bond; b) Reaction of CO₂with coordinated olefins.

commercially are discussed. The order of the discus- sion is structured according to the underlying reaction principle.

2.2 Oxidative coupling of CO₂and olefins – metallalactones as intermediates for the catalytic synthesis of acrylic acid

The oxidative coupling of CO₂and olefins is a very interesting chapter of academic CO₂ chemistry, and it also has an important direct link to industrial pro- duction technologies for acrylic acid. Acrylic acid is a starting material for the production of polyacrylates.

Currently acrylic acid is synthesized industrially by the SOHIO process on the 3×10⁶t/a scaleviathe oxida- tion of acroleine at heterogeneous molybdenum oxide- vanadium oxide catalysts at temperatures of 300^◦C.

As acroleine is produced by the oxidation of propylene oxide at bismuth-molybdenum oxide catalysts [60], it is obvious that the replacement of the high-energy two-step processes by a one-step low-energy pathway would be very interesting to enable a more sustainable overall route to acrylic acid. Given the increasing de- mand of propylene in the polyolefin industry, the use of ethylene as raw material would currently also be an attractive feature. Finally, a significant amount of CO₂could be fixed in the polyacrylates for long time frames.

The group of Aresta published in 1975 that CO₂re- acts with [Ni(cod)₂] (cod=cis,cis-1,5-cyclooctadiene) in the presence of 1,2-bis(dicyclohexylphosphanyl)- ethane (dcpe) yielding [(η²-CO₂)Ni(dcpe)] [61]. In the following years a lot of work was devoted to the ox- idative coupling of CO₂ with olefins at nickel com- plexes [38] by many groups, with detailed fundamen-

tal knowledge resulting from the systematic work by Hoberg [62]. The nickelalactones undergo hydrolysis in the presence of mineral acids leading to the degra- dation of the nickel complex and formation of the free acrylic acid in a stoichiometric reaction. In prin- ciple, one can envisage to assemble a catalytic cy- cle as depicted in Scheme2 on the basis of these ob- servations. In this mechanism, CO₂and ethylene can firstly be coupled to a metallalactone and subsequently be cleaved from the complex by β-hydride elimi- nation to yield the corresponding unsaturated com- pound acrylic acid and retain the catalytically ac- tive species. Owing to the thermodynamics, the free acrylic acid would have to be stabilized in the form of a salt or another derivative to provide a driving force.

However, the β-hydride elimination from nicke- lalactones proved to be difficult. The reason is the steric hindrance induced by the rigidity of the planar metallalactone ring which makes it difficult for theβ- hydrogen atom to come close enough to the nickel center. In a large number of experiments and theoreti- cal investigations by many different groups it could be shown during the past decades that theβ-hydride elim- ination for the production of acrylic acid from nick- elalactones is only possible under very specific con- straints preventing currently a direct route to acrylic acid.

Scheme 2. Hypothetical catalytic reaction of CO₂with ethy- lene to yield acrylic acid using nickel(0) catalysts.

Scheme 3. Step-wise synthesis of methyl acrylate from CO2, ethylene and methyl iodide at [(dppe)Ni] [64].

DFT computations have suggested that β-hydride elimination should be possible in principle when the Ni–O distance is extended, and in this way an agos- tic interaction between the nickel center and the β- hydrogen atom is enabled [63]. Accordingly, the open- ing of the metallalactone ring by breaking the Ni–

O bond should favor the β-hydride elimination. The group of Rieger showed very recently that these pre- dictions indeed allow for the formation of acrylic acid esters. They used the oxidative addition of MeI to open the dppp-complexed nickelalactone by oxidative addi- tion of MeI as shown in Scheme3[64]. The resulting open chain species is flexible enough to undergo theβ- hydride elimination, and methacrylate is produced as the thermodynamically favored product. The HI elimi- nation resulting in the Ni(0) species would close a cat- alytic cycle, but this species cannot be converted to the metallalactone complex under the given reaction con- ditions. Accordingly, the reaction is not yet catalytic, and also substantial amounts of MeI (100 fold excess) are needed to obtain significant amounts of acrylate.

In 2012 the groups of Limbach and Hofmann reported on the catalytic formation of sodium methacrylate using [(dtbpe)Ni] (dtbpe=1,2-bis-(di- tert-butylphosphanyl)ethane) in a two-step process for the reaction of CO₂and ethylene. They used sodium tert-butylat (NaOtBu) as a base to facilitate both the formal β-hydride elimination as well as the prod- uct stabilization [65]. The postulated reaction mech- anism for the overall reaction sequence is depicted in Scheme4.

Although the amount of product formed per nickel center is currently equivalent to only 10 catalytic turnovers, this provides the first evidence for a catalytic production of sodium acrylate by a homogeneous cata- lyst. As such it serves as a nice example of a fruitful interplay between industrial needs and fundamental re- search.

2.3 Insertion of CO₂into M–O bonds – synthesis of carbonates, polycarbonates and polyurethanes

Aliphatic and cyclic carbonates are valuable inter- mediate products in the chemical industry [66–70].

Dimethylcarbonate (DMC) for instance is used as a methylating agent in organic syntheses, for polymer fabrication and as a solvent and lubricant. The trans- esterifciation with phenol leads to diphenylcarbonate, which can be further reacted with Bisphenol A leading to the corresponding polycarbonate. Currently the re- action of phosgene with the corresponding alcohol is used for the synthesis of aliphatic carbonates [71], and alternatives based on more benign starting materials are the subject of many research efforts. The direct use of CO₂ remains a significant challenge, owing to the formation of water as the coupled by-product which imposes challenges on reaction engineering as well as catalyst design. Recent advances include the use

Scheme 4. Step-wise synthesis of sodium acrylate from ethy- lene and CO₂with [Ni(dtbpe)] as mediator, resulting overall in a catalytic coupling of CO2and ethylene [65].

Scheme 5. Synthesis of cyclic carbonates from epoxides and CO2.

of niobium-based systems described by Aresta for the synthesis of DMC and other carbonates [72–74]. Tin- and titanium-based mediators have been studied exten- sively by the groups of Ballivet-Tkatchenko [75–77]

and Sakakura [78,79], respectively.

Five-ring carbonates can be obtained by the reaction of CO₂with an epoxide in the presence of an appro- priate catalyst (Scheme5) [80–83]. The correspond- ing synthesis of ethylene and propylene carbonate is carried out on an industrial scale. Recent reviews by Sakakura [13,67] summarize the current state of the art in industry and academia. Transition metal cata- lysts have been under investigation for more than three decades, and an initial report about the catalytic syn- thesis of (poly)-carbonates was published 1978 by In- oue and Takeda [84]. Currently mono- and binuclear salen complexes of aluminum, chromium and cobalt are studied predominantly [85–88], aiming mainly at an improved activity and at new synthetic applications including a stereocontrol at the chiral center.

The currently accepted catalytic cycle for the for- mation of cyclic carbonates from epoxides and CO₂is depicted in Scheme6. The initial step is the coordina- tion of the epoxide viaits oxygen atom to the metal center of the catalyst, resulting in an activation of the epoxide. Secondly the epoxide ring is opened by assis- tance of a halide or another nucleophile (Nuc) yielding an alkoxide. Subsequently, CO₂inserts into the M–O bond, forming a carboxylate complex. Back-biting of the carboxylate under elimination of the nucleophile leads to the formation of the five-membered cycle. The intriguing interplay of Lewis-acid and Lewis-base ac- tivation is crucial for the development of bi-functional and bimetallic catalysts: One site binds the epoxide and the other coordinates CO₂, which then acts as a nucle- ophile and induces the ring opening of the epoxide.

Cyclic as well as open-chain aliphatic carbon- ates can be used as carriers for the carbonate unit in the synthesis of polycarbonates [89]. However,

O O

O

R²

R¹ X-MLn

O R¹

R²

O R¹

R² Ln M X

X Ln

M O Nuc

R² R¹ X L_n

M O O

O

R²

Nuc R¹

Nuc

CO₂

Scheme 6. Simplified mechanism for the formation of cyclic carbonates from CO2and epoxides. Nuc=nucleophile.

Scheme 7. Co-polymerization of epoxides and CO2.

polyalkylenecarbonates can also be obtained directly via catalytic co-polymerization of epoxides and CO₂ in high efficiency (Scheme7) [13,59]. Depending on the catalyst, the reaction can be controlled to pro- duce either polyalkylenecarbonates of high molecu- lar weight with fully alternating repetition units, or lower molecular weight materials with random incor- poration of carbonate linkages. High molecular weight alternating polypropylenecarbonate is investigated in- tensively as a new polymer and for polymer blends in academia and industry [90,91]. The shorter chain copolymers of propylene oxide and CO₂show promis- ing properties as polyol-building blocks in the pro- duction of polyurethanes, and pilot plant operations are currently under way at BayerMaterialScience in Leverkusen [92]. The development of the catalytic processes for the generation of the polyol building

blocks was facilitated by the close collaboration be- tween academia and industry at CAT Catalytic Centre, a joint research institution of RWTH Aachen Univer- sity and the Bayer company. This successful approach provides an illustrative example for the innovation po- tential when fundamental research is applied to cur- rent challenges of industry in an appropriate research structure.

Polyether-polycarbonate-polyols bearing terminal alcohol groups are particularly interesting, as they pro- vide entry to polyurethanes. Within the context of CO₂ utilization it is important to realize that the incor- poration of CO2 into these materials opens the way to industrial products which store CO₂for long time frames. As polyurethanes constitute a market ofca.14 million tons per year it becomes evident that a signif- icant amount of CO₂ can be incorporated into these products. Furthermore, novel materials can be devel- oped to extend the property portfolio of the exist- ing products and in this way enrich the variety of polyurethane applications.

2.4 Insertion of CO2into M–H bonds – synthesis of methanol and formic acid

2.4.1 Hydrogenation of CO₂to methanol

Methanol is produced on the 30 million ton scale [93] annually as one of the most important basic chemicals, as a widely used industrial sol- vent, and for fuel additives such as methyl tert- butylether (MTBE) [94]. Current industrial produc- tion of methanol is based on syngas (CO/H₂) which is mainly obtained from natural gas [95,96]. CO₂ is sometimes present in the gas feed of these pro- cesses to balance the CO/H₂ ratio, and this is cur- rently the second largest use of CO₂in industrial syn- thesis after the urea processes, converting approxi- mately 2 Mio t CO₂ per year. The direct synthesis of methanol by hydrogenation of pure CO₂is possi- ble in principle (Scheme8) using heterogeneous-type catalysts. This has stimulated various concepts to use CO₂-based methanol as a sustainable carbon-based en- ergy carrier and entry point into the chemical supply chain [11–13].

The use of CO₂ as an energy vector is closely re- lated to the availability of hydrogen from non-fossil resources. A recent analysis [59] of the carbon foot- print of the production process of methanol has shown

Scheme 8. Generation of methanol by hydrogenation of car- bon dioxide and carbon monoxide.

that a net emission of 0.24 tons of CO₂ per ton of methanol is generated using syngas from steam re- forming of natural gas. This value even increases dras- tically to 4.29 t(CO₂)/t(MeOH) when CO₂is used as carbon source together with H₂obtained by electroly- sis of water usinge. g.the current German power mix.

However, if hydrogen production relies on electroly- sis of water with electricity generated solely by re- newable energy sources then the net CO₂emission for methanol production would be negative, and 1.38 tons of CO₂per ton of methanol could be fixed. In particu- lar, electricity peaks from renewable power generators that cannot be accommodated in the grid might offer an attractive option to be coupled to CO₂hydrogenation.

The prerequisite to make any of these scenarios vi- able is the availability of efficient catalysts for the hy- drogenation of CO₂to methanol. Heterogeneous cat- alysts have long been known for this reaction, and recent publications demonstrate the continuing inter- est [97–99]. Heterogeneous systems are currently al- ready investigated for larger scale application [100]. In sharp contrast, no molecular catalyst has been known until recently to enable the multistep reduction se- quence required for the hydrogenation of CO2 to methanol. In groundbreaking studies, the group of Mil- stein reported ruthenium pincer compounds which suc- cessfully catalyze the hydrogenation of carbonic acid derivatives and even formates to methanol [101,102].

This opened the possibility to hydrogenate CO2 to methanol stepwise through such intermediates, if a cat- alyst could affect each of the steps. In a first ap- proach, the group of Sanford used a cascade system of three different catalysts and observed the formation of methanol, albeit with very low turnover numbers [103].

Most recently, the first example for a single- site organometallic catalyst has finally been re- ported which allows the direct hydrogenation of CO₂ to methanol (Scheme9) [104]. The catalyst precursor is a ruthenium triphos complex which bears the trimethylenemethane (TMM) ligand. Upon activation with catalytic amounts of acids with non-coordinating anions such as methanesulfonate or bis(trifluoromethanesulfon)amide (BTA) methanol

Scheme 9. Synthesis of methanol from CO2 catalyzed by a ruthenium triphos complex.

was formed under relatively mild conditions with turnover numbers up to 220. Studies on the hydro- genation of formate esters and isotopic labeling exper- iments have confirmed a reaction sequence involving formate species as key intermediates. This opens new possibilities for rational catalyst development based on the molecular approach of organometallic chemistry which complements the development of multi-site het- erogeneous catalysts in the synthesis of methanol from CO₂.

2.4.2 Hydrogenation of CO₂to formic acid

The initial step of the hydrogenation sequence described above reduces CO₂ to the formate level.

Formic acid itself is in fact also a very interesting target product with applications among others in agriculture (silage) and the leather industry, and as a commodity chemical. It is produced on the scale of several hundred thousand tons per year mainly by the reaction of carbon monoxide (CO) with methanol to yield the methylester of the acid [59]. Subsequently, the ester is cleaved with base or acid to yield the salts or free acid, respectively (Scheme10).

Alternatively, CO2could be transformed into formic acid directly by simple addition of H₂. However, the thermodynamics of the reaction impose a major con- straint on the feasibility of the reaction. The forma- tion of formic acid from CO₂and H₂is an exothermic

Scheme 10. Step-wise industrial synthesis of formic acid by carbonylation of methanol and hydrolysis.

Scheme 11. Reaction of carbon dioxide with hydrogen to formic acid and its stabilization by formation of the amine adduct.

(∆H= −31.6 kJ mol⁻¹), though endergonic process (∆G= +32.9 kJ mol⁻¹), which makes the direct syn- thesis impossible [15]. To shift the equilibrium towards isolable amounts of the reduced species, the addition of appropriate stabilizers such as alkylamines or sodium hydroxide is necessary to form corresponding adducts or formate salts, thus establishing both exothermic and exoergic reactions (Scheme11).

Research for the development of catalysts for this reaction dates back to the mid 1970’s, and nowadays there is a large variety of primarily rhodium, ruthe- nium, iridium and even iron catalysts with in some cases very high turnover frequencies and turnover numbers [58,59]. The catalytic reaction relies on tran- sition metal hydrides as catalytically active species, and the currently accepted reaction mechanism is de- picted in Scheme12.

The details of the hydride attack at the carbon cen- ter of CO₂ during the formal insertion and the hy- drolytic cleavage of the resulting metal-oxygen bond depend strongly on the nature of the catalyst. For pro- totypical rhodium catalysts [105], the CO₂ molecule inserts into the M–H bond without the need for sta- ble pre-coordination. Activation of the H–H bond dur- ing hydrogenolysis can occur either by oxidative ad- dition orσ-bond metathesis, the latter pathway being preferred according to computational studies. Kinetic as well as computational studies consistently indicate

Scheme 12. General reaction mechanism for the catalytic hy- drogenation of CO2to formic acid or formates.

that the rate-limiting barriers are in the later steps of the mechanism liberating the product and regenerating the active species. As this step is associated with the thermodynamically unfavorable formation of the free acid, the base is important to improve yields and rates of the reaction by deprotonation of the acid as soon as it is generated.

Although a large number of catalytic systems with high activity and productivity have been described, the above described interplay between catalyst and stabi- lizer prevents the production of the desired free acid in a simple way. Multi-step processes have been de- veloped to enable the separation of the adduct from the catalyst and subsequent isolation of the free acid [106].

Several studies describe processes in which the adduct of formic acid and NEt₃is transformed into a thermally cleavable salt by exchange of the base [107–110]. Fur- thermore, multiphase reaction systems were used to in- vestigate the separation of the adduct from the catalyst.

Very recently a novel process was introduced by BASF in which trihexylamine was used as the base together with a ruthenium catalyst in protic media [106,110].

The corresponding adduct formed from formid acid and the amine could be separated form the aqueous phase and cleaved by distillation. Also immobilized ruthenium catalysts were used by the group of Zhang in combination wih ionic liquids (IL) carrying tertiary amino groups as basic functions and water [111,112].

The chemical transformation took place at the solid catalyst that was removed by filtration to obtain an IL phase which contained the product. This phase was subsequently distilled which led to a cleavage of the acid/base adduct, and free formic acid was obtained.

Most recently, the first example of a fully integrated process scheme has been presented allowing produc- tion of free formic acid directly in a single process unit [113]. The concept relies on a continuous-flow process scheme where carbon dioxide plays a dual role as the reactant and as the mobile phase in the supercrit- ical state (scCO₂). As depicted in Fig.2, the catalyst and the stabilizing base are contained in the stationary phase consisting of an ionic liquid (IL) which is not miscible with scCO₂.

During the reaction the scCO₂/H₂ phase moves through the reactor, and CO₂ as well as H₂ can dif- fuse into the IL phase where they react to formic acid at the catalyst. The solid base present in the IL phase stabilizes the formic acid in form of the typical for- mate adduct. The flow of scCO₂is able to extract part

Fig. 2. Integrated process scheme for the catalytic hydro- genation of CO2to free formic acid under continouos flow conditions using scCO2 as the extractive mobile phase and a stationary IL phase containing the catalyst and the base.

of the formic acid from the reaction mixture provided that its vapor pressure in the base adduct is sufficiently large to make it soluble in the scCO₂phase. Stable per- formance over more than 200 hours on stream was ob- served for selected combinations of catalyst, IL, and base. This illustrates how rational process design can help to overcome limitations encountered in batch- wise operation through continuous-flow systems based on advanced fluids as reaction and separation media.

2.5 Insertion of CO₂into the M–C bond – direct synthesis of aromatic carboxylic acids

The stoichiometric reaction of CO2with alkali phe- nolates yielding hydroxy carboxylic acids dates back to the 1870’s [114–120]. Today the process is known as the Kolbe-Schmitt reaction, and either salicylic acid orp-hydroxy benzoic acid can be obtained selectively by choosing either sodium or potassium ions, respec- tively, to balance the anionic charge of the phenolate (Scheme13). Theorthoselectivity in the presence of sodium (and also lithium) ions is a result of the chelate effect exerted by these cations [121], while the larger potassium ions drive the reaction selectively towards thepara-substituted product [122,123]. The synthesis of pharmaceuticals like acetyl salicylic acid (ASS, As- pirin), and of dyes and pesticides are the main applica- tions for salicylic acid.

In general, aromatic carboxylic acids are important structural motifs in many pharmaceuticals, agrochem- icals, and other fine chemicals and specialty products.

The development of catalytic methods for the formal insertion of CO₂into the C–H bonds of arenes would

Scheme 13. Synthesis ofp-hydroxybenzoic acid and salicylic acid, and further conversion to acetyl salicylic acid (ASS, Aspirin).

thus open a large field of possible synthetic applica- tions. Despite significant recent progress this field is still in its infancy, and further fundamental academic research is needed to develop efficient systems. The insertion of CO₂into the C–H bond of aromatic com- pounds suffers from the same thermodynamic limita- tions as the hydrogenation of CO₂to formic acid mak- ing the utilization of an appropriate base inevitable.

The carboxylation of benzene as a prototype reaction with the associated reaction enthalpy and the Gibbs free reaction energy is shown in Scheme14.

Current synthetic techniques by which the CO₂ molecule can be introduced into arenes involve clas- sical lithium or Grignard aryl compounds, aryl boronic esters [124,125] or aryl halides [126,127], which

Scheme 14. Synthesis of benzoic acid from benzene and CO₂ as a prototypical example for the direct carboxylation of arenes. Reaction energy and reaction Gibbs free energy in kcal mol⁻¹.

Scheme 15. Catalytic cycle of the insertion of CO2into the C–H bonds of arenes using gold and copper NHC catalysts.

makes the carboxylation an overall multistep process.

An intriguing example for direct carboxylation of aro- matic C–H bonds was presented recently by the group of Nolan who introduced gold hydroxyN-heterocyclic carbene (NHC) complexes as catalysts for the reac- tion of CO₂with electron-deficient arenes [128–130].

Shortly after Nolan’s report the group of Hou described similar insertions of CO₂into such arenes using cop- per chloride NHC complexes [131]. The postulated re- action mechanism is shown in Scheme15.

In the first step an acidic C–H group of the substrate reacts with the hydroxy group of the NHC complex lib- erating water. CO₂subsequently inserts into the M–C bond with formation of the corresponding carboxylate, which is cleaved from the complex as its potassium salt regenerating the M–OH group at the metal complex.

As stated above, the reaction relies on the acidity of the C–H bond, which is cleaved in the initial step and needs pK_Avalues of<32, presenting a clear limitation in substrate scope.

The group of Iwasawa reported on the rhodium- catalyzed carboxylation of arenes with CO₂ initiated by directed metalation of substrates like 2-phenyl- pyridine or 1-phenylpyrazole [132]. Again, the inser- tion of CO₂into the resulting Rh–C bond is believed to give the carboxylate intermediate. Regeneration of the active species by Rh–O cleavage and stabiliza- tion of the products in form of the thermodynami-

cally favored esters requires two equivalents of alumi- nomethyl reagent per equivalent of substrate.

In order to assess systematically the various con- trol factors that influence a potential catalytic cycle, a computational study investigating the potential of a series of ruthenium pincer complexes as catalysts in the direct carboxylation of benzene and related arenes was undertaken [133]. The study was based on a gen- erally applicable catalytic cycle in which no special requirements as to the properties of the arene were made. The following assumptions guided the design of the catalytic cycle: Ruthenium phosphine complexes with coordinated benzoate ligands are available syn- thetically and have been characterized crystallograph- ically. Such a benzoate complex should inevitably be present at some stage of the catalytic cycle if the for- mation of a C–C bond between CO₂and a metal-bound arene does occurvia an insertion mechanism. Ruthe- nium pincer complexes are widely known, and the typ- ically observed meridional coordination mode allows within certain limits for a systematic variation of the electronic properties of the metal center. The starting point of the hypothetical catalytic cycle was therefore defined with a benzoate ligand coordinated bidentately at the metal center, while the third coordination site in the same plane is occupied by an anionic spectator lig- and. The pincer ligand adopts the remaining three co-

Scheme 16. Postulated general mechanism for the direct car- boxylation of benzene with CO₂using ruthenium pincer cat- alysts. X=monodentate anionic ligand.

ordinations sites of an octahedrally coordinated ruthe- nium(II) center (Scheme16).

The first step of the catalytic process requires to open a coordination site by change from a bidentately coordinated benzoate to a monodentately coordinating one. The vacant coordination site is then filled by an incoming arene, which coordinates to the metalviathe C–H bond. The first bond breaking/bond forming event consists of the H transfer from the arene in aσ-bond metathesis to a carboxylate oxygen atom. The carbon atom of the arene ring forms a covalent bond to the metal center, while the Ru–O linkage is changed from covalent to a simply coordinating mode. In the next step the trialkylamine base cleaves the benzoic acid from the complex and forms the typical ammonium benzoate adduct as product. The vacant coordination site at the metal is occupied at this stage by an incom- ing CO2molecule, which subsequently inserts into the Ru–C bond and in this way regenerates the catalyti- cally active species.

Notably, the proposed catalytic cycle could be suc- cessfully modeled by location of all intermediates and transition states in a computational screening process for a large number of different ruthenium pincer cat- alysts. Predicted turnover frequencies were estimated following the analysis introduced recently by the group of Shaik [134–136] from computed catalytic cycles, allowing a detailed assessment of the relative changes in reaction rates associated with variations of individ- ual components and parameters of the catalytic sys- tem. Based on these results, it was possible to identify catalyst lead structures which are predicted to gener- ate TOFs high enough to yield benzoic acid deriva- tives under mild conditions. These results are ex- pected to stimulate experimental work to synthesize such catalysts in order to validate the computational predictions.

3. Conclusion and Outlook

This short review highlighted current scientific de- velopments in the utilization of CO₂as a C1 building block for the industrial synthesis of chemical products.

The following general conclusions can be drawn for the individual fields of potential application:

⇒Very close to industrial realization or even in the state of becoming commercialized is the copolymer- ization of epoxides with CO₂in the field of polycar-

bonate and polyurethane fabrication. The introduc- tion of these new technologies into industrial pro- cesses would open a new chapter in the use of CO₂ as industrial feedstock on a large scale, resulting in improved carbon footprints for the production pro- cesses of these materials.

⇒A significant impact on the carbon balance of the chemical and energetic supply chains could result from the combination of renewable hydrogen pro- duction with CO₂ reduction. For the CO₂-based production of methanol the development of a new catalyst technology is under way, and novel het- erogeneous and homogeneous catalysts have pro- vided new opportunities in this field. The synthe- sis of formic acid has reached a state where a large number of efficient catalysts are now available. The next level of innovation is expected to come from a process technology addressing the challenges of product separation and isolation resulting from the thermodynamic constraints of the transformation.

⇒The direct catalytic carboxylation or aromatic sub- strates can offer a widely useful synthetic strategy (“catalytic Kolbe-Schmitt chemistry”), and first at- tempts to identify lead structures for catalysis de-

velopment are providing promising results. How- ever, large efforts in fundamental research are still required to arrive at synthetically useful and reliable methods of general applicability.

It is important to note that the rapid progress in this field in the last few years was based on three main pil- lars:

1) Strong dynamics initiated by significant funding schemes targeting the area in a strategic way.

2) Close interaction between academic research and in- dustrial drivers brought together by open forms of col- laboration in appropriate formats.

3) A strong basis in fundamental research in this field which has developed over decades, advancing largely by curiosity-driven and visionary endeavors.

Obviously, points 1 and 2 would not have been pos- sible without the foundation of point 3. This should be taken as an important message to ensure sufficient room for fundamental academic research in modern science policy and funding, as it provides the grounds for technological and industrial innovation and will continue to lay the basis of sustainable solutions for current and future global challenges.

[1] BP Statistical Review of Global Energy, 2012;

www.bp.com/statisticalreview (retrieved August 31, 2012).

[2] F . Sch¨uth inDie Zukunft der Energie, (Eds.: P. Gruss, F. Sch¨uth), C. H. Beck, M¨unchen,2008, pp. 15 – 31.

[3] P. Markewitz, W. Kuckshinrichs, W. Leitner, J. Lins- sen, P. Zapp, R. Bongartz, A. Schreiber, Th. E. M¨uller, Energy Environ. Sci.2012,5, 7281 – 7305.

[4] M. Aresta, Carbon Dioxide as Chemical Feedstock, Wiley-VCH, Weinheim,2010.

[5] W. Kuckshinrichs, P. Markewitz, J. Linssen, P. Zapp, M. Peters, B. K¨ohler, Th. E. M¨uller, W. Leitner, Weltweite Innovationen bei der Entwicklung von CCS-Technologien und M¨oglichkeiten der Nutzung und des Recyclings von CO2, Schriften des Forschungszentrums J¨ulich, Vol. 60, J¨ulich 2009;

ISBN 978-3-89336-617-0.

[6] P. Heffer, M. Prud’homme, 79^thIFA Annual Confer- ence, Montreal (Canada), 23 – 25 May2011.

[7] G. Pagani,Hydrocarbon Process. Int. Ed.1982, 61, 87 – 91.

[8] J. A. Ritter, A. D. Ebner,Sp. Sci. Technol.2007, 42, 1123 – 1193.

[9] M. Ahmadinejad, M. R. Desai, T. C. Watling, A. P. E.

York,Adv. Chem. Eng.2007,33, 47 – 101.

[10] F. Asinger,Methanol – Chemie und Energierohstoff.

Die Mobilisation der Kohle, Springer, Berlin1986.

[11] W. Keim, H. Offermanns, Nachr. Chem. 2010, 58, 434 – 435.

[12] A. Behr,Carbon Dioxide Activation by Metal Com- plexes, Wiley-VCH, Weinheim1988.

[13] T. Sakakura, J. Choi, H. Yasuda, Chem. Rev. 2007, 107, 2365 – 2387.

[14] W. Leitner,Coord. Chem. Rev.1996,153, 257 – 284.

[15] W. Leitner, Angew. Chem., Int. Ed. Engl.1995, 34, 2207 – 2221.

[16] W. Keim, A. Behr, H. Rzehak, Tenside 1979, 3, 113 – 116.

[17] J. P. Bianchini, B. Wegell, E. Gaydou, H. Rzehak, W. Keim,J. Mol. Catal.1981,10, 247 – 252.

[18] A. Behr, W. Keim, G. Thelen,J. Organomet. Chem.

1983,249, C38–C40.

[19] A. Behr, E. Herdtweck, W. A. Herrmann, W. Keim, W. Kipshagen,Chem. Commun.1986, 4263.

[20] A. Behr, K. D. Juszak, W. Keim, Synthesis 1983, 574.

[21] D. Vogt, Habilitation Thesis, RWTH Aachen, Aachen 1998.

[22] A. Behr, G. Henze,Green Chem.2011,13, 25 – 39.

[23] A. Behr, M. Becker, T. Beckmann, L. Johnen, J. Le- schinski, S. Reyer,Angew. Chem. Int. Ed. 2009, 48, 3598 – 3614.

[24] A. Behr, Angew. Chem., Int. Ed. Engl. 1988, 27, 661 – 678.

[25] A. Behr, K.-D. Juszak, W. Keim,Synthesis1983, 574.

[26] A. Behr, K.-D. Juszak, J. Organomet. Chem. 1983, 255, 263 – 268.

[27] A. Behr, R. He, K.-D. Kuszak, C. Kr¨uger, Y.-H. Tsay, Chem. Ber.1986,119, 991 – 1015.

[28] H. Hoberg, S. Schaefer, J. Organomet. Chem.1982, 236, C28–C30.

[29] H. Hoberg, S. Schaefer, J. Organomet. Chem.1983, 251, C51–C53.

[30] H. Hoberg, A. Ballestros, A. Sigan, C. Jegat, A. Mil- chereit,Synthesis1991, 395 – 398.

[31] H. Hoberg, S. Schaefer, G. Burkhart, C. Kr¨uger, M. J. Rom˜ao, J. Organomet. Chem. 1984, 266, 203 – 224.

[32] H. Hoberg, D. Schaefer, B. W. Oster, J. Organomet.

Chem.1984,266, 313 – 320.

[33] H. Hoberg, K. Jenni, C. Kr¨uger, E. Raabe, Angew.

Chem., Int. Ed. Engl.1986,25, 810 – 811.

[34] H. Hoberg, K. Jenni,J. Organomet. Chem.1987,322, 193 – 201.

[35] H. Hoberg, K. Jenni, K. Angermund, C. Kr¨uger, Angew. Chem., Int. Ed. Engl.1987,26, 153 – 155.

[36] D. Walther,Coord. Chem. Rev.1987,79, 135 – 174.

[37] D. Walther, G. Br¨aunlich, R. Kempe, J. Sieler, J.

Organomet. Chem.1992,436, 109 – 119.

[38] D. Walther, H. Sch¨onberg, E. Dinjus, J. Organomet.

Chem.1987,334, 377 – 388.

[39] D. Walther, E. Dinjus, J. Sieler, L. Andersen, O. Lind- qvist,J. Organomet. Chem.1984, 395 – 398.

[40] E. Dinjus, J. Kaiser, J. Sieler, D. Walther, Z. Anorg.

Allg. Chem.1981,483, 63 – 68.

[41] J. Kaiser, J. Sieler, U. Braun, L. Golic, E. Dinjus, D. Walther, J. Organomet. Chem. 1982, 224, 81 – 87.

[42] D. Walther, E. Dinjus, H. G¨orls, J. Sieler, O. Lind- qvist, L. Andersen,J. Organomet. Chem.1985,286, 103 – 114.

[43] Y. Sakaki, Y. Inoue, H. Hashimoto, J. Chem. Soc., Chem. Commun.1976, 605 – 606.

[44] Y. Inoue, Y. Sasaki, H. Hashimoto,Bull. Chem. Soc.

Jpn.1978,51, 2375 – 2378.

[45] Y. Inoue, Y. Itho, H. Hashimoto,Bull. Chem. Soc. Jpn.

1980,53, 3320 – 3333.

[46] Y. Fukue, S. Oi, Y. Inoue,J. Chem. Soc., Chem. Com- mun.1994, 2091.

[47] Y. Inoue, H. Izumida, Y. Sasaki, H. Hashimoto,Chem.

Lett.1976, 863 – 864.

[48] A. Musco, C. Perego, V. Tartiari, Inorg. Chim. Acta 1978,28, L147–L148.

[49] T. Yamamoto, K. Igarashi, S. Komiya, A. Yamamoto, J. Am. Chem. Soc.1980,102, 7448 – 7456.

[50] S. Saito, S. Nakagawa, T. Koizumi, K. Hirayama, Y. Yamamoto,J. Org. Chem.1999,64, 3975 – 3978.

[51] D. J. Darensbourg, M. W. Holtkamp, Coord. Chem.

Rev.1996,153, 155 – 174.

[52] D. J. Darensbourg, Inorg. Chem. 2010, 49, 10765 – 10780.

[53] D. A. Palmer, R. van Eldik, Chem. Rev. 1983, 83, 651 – 731.

[54] D. Gibson,Chem. Rev.1996,96, 2063 – 2095.

[55] D. H. Gibson, Coord. Chem. Rev. 1999, 185 – 186, 335 – 355.

[56] X. Lin, J. R. Moss, Coord. Chem. Rev. 1999, 181, 27 – 59.

[57] C. Song,Catal. Today2006,115, 2 – 32.

[58] M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herr- mann, F. E. K¨uhn, Angew. Chem. Int. Ed.2011,50, 8510 – 8537.

[59] M. Peters, B. K¨ohler, W. Kuckshinrichs, W. Leitner, P. Markewitz, Th. E. M¨uller,ChemSusChem2011,4, 1216 – 1240.

[60] W. A. Herrmann in Applied Homogeneous Cataly- sis with Organometallic Compounds, Vol. 3 (Eds.:

B. Cornils, W. A. herrmann), 2^nd ed., Wiley-VCH 2002, pp. 1297 – 1304.

[61] M. Aresta, F. Nobile, V. G. Albano, E. Forni, M. Ma- nasero, J. Chem. Soc., Chem. Commun. 1975, 636 – 637.

[62] G. Burkhart, H. Hoberg,Angew. Chem., Int. Ed. Engl.

1982,21, 76.

[63] M. Aresta, C. Pastore, P. Giannoccaro, G. Kovacs, A. Dibenedetto, I. Papai, Chem. Eur. J. 2007, 13, 9028 – 9034.

[64] C. Bruckmeier, M. W. Lehenmeier, R. Reichhardt, S. Vagin, B. Rieger, Organometallics 2010, 29, 2199 – 2202.

[65] M. L. Lejkowski, R. Lindner, T. Kageyama, G. E. Bodiz, P. N. Plessow, I. B. M¨uller, A. Sch¨afer, F. Rominger, P. Hofmann, C. Futter, S. A Schunk, M. Limbach, Chem. Eur. J. 2012, doi:10.1002/chem.201201757.

[66] D. Ballivet-Tkachenko, S. Sorokina inCarbon Diox- ide Recovery and Utilization(Ed.: M. Aresta), Kluwer Academic Publishers, Dordrecht2003, 261 – 277.

[67] T. Sakakura, K. Kohno, Chem. Commun. 2009, 1312 – 1330.

[68] F. Arico, P. Tundo, Russ. Chem. Rev. 2010, 79, 479 – 489.

[69] T. Sakakura, J.-C. Choi, H. Yasuda, T. Takahashi, K. Kohno, S. Kitazume, L.-N. He,Prepr. - Am. Chem.

Soc., Div. Pet. Chem.2010,55, 49.

[70] S. Fujita, B. M. Bhanage, M. Arai in Progress in Catalysis Research(Ed.: L. P. Bevy), Nova Science, New York2005, pp. 57 – 79.

[71] R. Zevenhoven, S. Eloneva, S. Teir, Catal. Today 2006,115, 73 – 79.

[72] M. Aresta, A. Dibenedetto, C. Pastore,Inorg. Chem.

2003,42, 3256 – 3261.

[73] M. Aresta, A. Dibenedetto, L. Gianfrat, C. Pastore, Appl. Catal. A2003,255, 5 – 11.

[74] M. Aresta, A. Dibenedetto, F. Nocito, C. Pastore, J.

Mol. Catal. A2006,257, 149 – 153.

[75] D. Ballivet-Tkatchenko, S. Chambrey, R. Keiski, R. Ligabue, L. Plasseraud, P. Richard, M. Turunen, Catal. Today2006,115, 80 – 87.

[76] M. P. Kalhor, H. Chermette, S. Chambrey, D. Ballivet- Tkatchenko, Phys. Chem. Chem. Phys. 2011, 13, 2401 – 2408.

[77] L. Plasseraud. D. Ballivet-Tkatchenko, H. Cattey, S. Chambrey, R. Ligabue, P. Richard, R. Willem, M. Biesemans, J. Organomet. Chem. 2010, 695, 1618 – 1626.

[78] K. Kohno, J.-C. Choi, Y. Ohshima, A. Yili, H. Ya- suda, T. Sakakura,J. Organomet. Chem. 2008,693, 1389 – 1392.

[79] K. Kohno, J.-C. Choi, Y. Ohshima, H. Yasuda, T. Sakakura,ChemSusChem2008,1, 186 – 188.

[80] T. Sakakura, K. Kohno, Chem. Commun. 2009, 11, 1312 – 1330.

[81] T. Takahashi, T. Watahiki, S. Kitazume, H. Yasuda, T. Sakakura,Chem. Commun.2006,15, 1664 – 1666.

[82] J. E. Dengler, M. W. Lehenmeier, S. Klaus, C. E. An- derson, E. Herdtweck, B. Rieger,Eur. J. Inorg. Chem.

2011, 336 – 343.

[83] S. Liang, H. Liu, S. J. Y. G. B. Han,Chem. Commun.

2011,47, 2131 – 2133.

[84] N. Takeda, S. Inoue, Makromol. Chem. 1978, 179, 1377 – 1381.

[85] A. Decortes, A. M. Castilla, A. W. Kleij, Angew.

Chem. Int. Ed.2010,49, 9822 – 9837.

[86] L.-N. He, J.-Q. Wang, J.-L. Wang,Pure Appl. Chem.

2009,81, 2069, 2080.

[87] J. Melendez, M. North, P. Villuendas, Chem. Com- mun.2009, 2577 – 2579.

[88] W. Clegg, R. W. Harrington, M. North, R. Pasquale, Chem. Eur. J.2010,16, 6828 – 6843.

[89] R. Schwalm, P. Ball, H. Fullmann, W. Heitz, Polym.

Prepr. Am. Chem. Soc. Div. Polym. Chem.1984,25, 272 – 273.

[90] www.novomer.com(retrieved August 31, 2012).

[91] www.basf.com(retrieved August 31, 2012).

[92] W. Leitner, T. E. M¨uller, C. G¨urtler,Bayer Magazine, Annual Report 2010(Ed.: J. Sch¨afer); Bayer AG, Lev- erkusen,2011, pp. 32 – 37.

[93] H.-J. Althaus, R. Hischier, M. Osses, A. Primas, S. Hellweg, N. Jungbluth, M. Chudacoff, Life Cycle Inventories of Chemicals, Swiss Center for Life Cycle Inventories, D¨ubendorf,2007.

[94] T. Mokrani, M. Scurrell, Catal. Rev. Sci. Eng.2009, 51, 1 – 145.

[95] K. Weissermel, H.-J. Arpe,Industrial Organic Chem- istry, Wiley-VCH, Weinheim2003, pp. 30 – 36.

[96] G. Olah, A. Coeppert, G. K. S. Prakash,Beyond Oil and Gas: The Methanol Economy, Wiley-VCH, Wein- heim2006.

[97] L. C. Grabow, M. Mavrikakis, ACS Catal. 2011, 1, 365 – 384.

[98] W. Wang, S. Wang, X. Ma, J. Gong,Chem. Soc. Rev.

2011,40, 3703 – 3727.

[99] F. Liao, Y. Huang, J. Ge, W. Zheng, K. Tedsree, P. Collier, X. Hong, S. C. Tsang, Angew. Chem. Int.

Ed.2011,48, 2162 – 2165.

[100] www.mitsuichem.com(retrieved August 31, 2012).

[101] E. Balaraman, C. Gunanathan, J. Zhang, L. J. W. Shi- mon, D. Milstein,Nat. Chem.2011,3, 609 – 614.

[102] E. Balaraman, Y. Ben-David, D. Milstein, Angew.

Chem. Int. Ed.2011,50, 11702 – 11705.

[103] C. A. Huff, M. S. Sanford,J. Am. Chem. Soc. 2011, 133, 18122 – 18125.

[104] S. Wesselbaum, T. v. Stein, J. Klankermayer, W. Leit- ner,Angew. Chem. Int. Ed.2012,51, 7499 – 7502.

[105] F. Hutschka, A. Dedieu, W. Leitner,Angew. Chem. Int.

Ed. Engl.1995,34, 1742 – 1745.

[106] T. Schaub, R. A. Paciello, Angew. Chem. Int. Ed.

2011,50, 7278 – 7282.

[107] D. Preti, C. Resta, S. Squarcialupi, G. Fachinetti, Angew. Chem. Int. Ed.2011,50, 12551 – 12554.

[108] J. J. Anderson, D. J. Drury, J. E. Hamlin, A. G. Kent, WO 8602066,1986.

[109] M. J. Green, A. R. Lucy, M. Kitson, S. J. Smith, EP 0329337.

[110] T. Schaub, R. Paciello, K.-D. Mohl, D. Schneider, M. Sch¨afer, S. Rittinger, WO 2010149507,2010.

[111] Z. Zhang, Y. Xie, W. Li, S. Hu, J. Song, T. Jiang, B. Han, Angew. Chem. Int. Ed. 2008, 47, 1127 – 1129.

[112] Z. Zhang, S. Hu, J. Song, W. Li, G. Yang, B. Han, ChemSusChem2009,2, 234 – 238.

[113] S. Wesselbaum, U. Hintermair, W. Leitner, Angew.

Chem. Int. Ed.2012,51, 8585 – 8588.

[114] S. Walendy, G. Franci`o, M. H¨olscher, W. Leitner in Exp. Green Sustainable Chem., (Eds.: H. W. Roesky, D. K. Kennepohl), Wiley-VCH, Weinheim 2009, pp.

275 – 277.

[115] Y. Kosugi, Y. Imaoka, F. Gotoh, M. A. Rahim, Y. Mat- sui, K. Sakanishi, Org. Biomol. Chem. 2003, 1, 817 – 821.

[116] H. Kolbe, E. Lautemann,Liebigs Ann. Chem.1860, 113, 125 – 127.

[117] H. Kolbe, E. Lautemann,Liebigs Ann. Chem.1860, 115, 157 – 206.

[118] H. Kolbe, E. Lautemann, Liebigs Ann. Chem.1860, 115, 178.

[119] R. Schmitt,German Patent 29939,1884, 233.

[120] R. Schmitt,J. Prakt. Chem.1885,31, 397.

[121] J. Markovic, J. P. Engelbrecht, S. Markovic,Z. Natur- forsch. A: Phys. Sci.2002,57, 812 – 818.

[122] S. E. Hunt, J. I. Jones, A. S. Lindsey, D. C. Killoh, H. S. Turner,J. Chem. Soc.1958, 3152 – 3160.

[123] Z. Markovic, S. Markovic, N. Manojlovic, J. Pre- dojevic-Simovic, J. Chem. Inf. Model. 2007, 47, 1520 – 1525.

[124] K. Ukai, M. Aoki, J. Takaya, N. Iwasawa, J. Am.

Chem. Soc.2006,128, 8706 – 8707.

[125] T. Ohishi, M. Nishiura, Z. Hou,Angew. Chem. Int. Ed.

2008,47, 5792 – 5795.

[126] A. Correa, R. Mart´ın,J. Am. Chem. Soc.2009, 131, 15974 – 15975.

[127] A. Correa, R. Mart´ın,Angew. Chem. Int. Ed.2009,48, 6201 – 6204.

[128] I. I. F. Boogaerts, S. P. Nolan,J. Am. Chem. Soc.2010, 132, 8858 – 8859.

[129] I. I. F. Boogaerts, G. C. Fortman, M. R. L. Furst, C. S. J. Cazin, S. P. Nolan, Angew. Chem. Int. Ed.

2010,49, 8674 – 8677.

[130] I. I. F. Boogaerts, S. P. Nolan,Chem. Commun.2011, 47, 3021 – 3024.

[131] L. Zhang, J. Cheng, T. Ohishi, Z. Hou,Angew. Chem.

Int. Ed.2010,49, 8670 – 8673.

[132] H. Mizuno, J. Takaya, N. Iwasawa,J. Am. Chem. Soc.

2011,133, 1251 – 1253.

[133] A. Uhe, M. H¨olscher, W. Leitner,Chem. Eur. J.2012, 18, 170 – 177.

[134] S. Kozuch, S. Shaik, J. Am. Chem. Soc. 2006, 128, 3355 – 3365.

[135] S. Kozuch, S. Shaik, J. Phys. Chem. A 2008, 112, 6032 – 6041.

[136] A. Uhe, S. Kozuch, S. Shaik,J. Comput. Chem.2011, 32, 978 – 985.